Using the process app for BAW (IBM Business Automation Workflow)

Setting up the BAW connector

The BAW connector is a Python module that is designed to extract raw data from Business Automation Workflow (BAW) and create CSV files. The connector offers interactive file-based options to configure the extraction of data.

This connector uses asynchronous programming to process multiple tasks concurrently and includes a scheduling feature for up to 24 hours. The data flow involves steps like gathering configuration details, setting logging, choosing execution modes, retrieving instance lists, extracting data in parallel, and generating CSV files.

To set up and use the BAW connector, complete the following steps:

1. Go to the BAW-Connector Directory.

   If you download the BAW-IBM-Process-Mining-Assets.zip file, extract the file and go to the BAW-Connector directory.

   If you download the BAW-Connector.zip file from the imported BAW Process App within Process Mining, extract the zip in BAW-Connector folder and access the BAW-Connector directory.

   cd <path to your folder>/BAW-Connector
   

   
   

2. Create a BAW Connector Virtual Environment.

   Use the Python virtual environment tool to create a dedicated environment for the BAW Connector.

   python -m venv bawc_venv
   

   
   

3. Activate the Virtual Environment

   Activate the BAW Connector virtual environment to isolate dependencies and ensure compatibility.

   • macOS/Linux

     source bawc_venv/bin/activate
     

   • Windows in CMD

     .\bawc_venv\Scripts\activate.bat
     

   • Windows in Powershell

     .\bawc_venv\Scripts\Activate.ps1
     

4. Install required dependencies.

Use pip to install all required dependencies/libraries listed in the requirements.txt file within the virtual environment.

pip install -r requirements.txt

Running the BAW Connector Module

To run the BAW connector module, you can use the following functions as entry points to the BAW_Connector.py module:

• main function (User interaction by command line)
• system_extract_baw_data function(interactive or system integration/automation

    python BAW_Connector.py ./baw_config.json

        {
        
        "root_url": "",
        
        "process_name": "",

        "project": "",

        "from_date": "2011-08-04T03:12:53Z",

        "from_date_criteria": "",

        "to_date": "2024-08-04T03:12:53Z",

        "to_date_criteria": "",

        "status_filter": "Active,Completed,Failed,Terminated,Suspended,Late,At_Risk",

        "instance_limit": 10,

        "offset": 0,

        "thread_count": 1,

        "csvpath": "",

        "csvfilename": "",

        "logfile": "",

        "user_name": "",

        "password": ""

        }

from main.BAW_Connector import system_extract_baw_data

config = {

...

    }

if name == "main":

system_extract_baw_data(config)

Uploading the custom process app to IBM Process Mining

After using the BAW Connector to generate your CSV file, you can use the BAW Process App to extract, transform, and load the process data.

The BAW Process App and Connector supports the BAW version 23.0.2.

To access the Process App services for BAW, do the following steps:

1. From the home page, click Start from a process app.

2. In the Process App page, click Import Process App from file.

3. Upload the BAW Business Process Management.pma file in the Process App dialog box and click Create. The Process App is created.

   

Creating a process by using the BAW process management app

To create the process app, complete the following steps:

1. In the Process applications page, click Get Started on the BAW process management tile.

2. In the Process details step, complete the following steps:

   a. Enter the name of your process.

   b. Optional: Link your process to an organization.

   • If you want to link your process to an existing organization, select the Use existing organization option and select an organization name from the list.

   • If you want to link your process to a new organization, select the Create new organization option and enter the name of the new organization.

   c. Click Next to go to the Documentation step.

   To gather data from the BAW Process App, you must download and configure the BAW connector/extractor code script to extract the BAW raw data from the BAW application. After you execute these steps, click Next to go to the Configuring event logs step.

   If you did not execute these steps, then select Help me gather my data.

   Click Download .gzfile to access the Python script to extract process data from a BAW application.

   After the execution is complete, a .gz file (which contains a CSV file of the BAW process data from the BAW application) is created in the location that you specify during that configuration and execution of the connector/extractor script. The .gz file is needed to proceed to the next step. After you have the .gz file, click Next to go to the Configuring event logs step.

3. Configuring Event Logs

   In the configuration event logs, you can add any process or task data into the CSV.

   In the Configuring Event Logs, select Include exposed variables checkbox to include the exposed process variables. The field name of each tracked variable in the Process Mining data source starts with 'trkd'.

   For example,

   trkd.requisition.requester
   

   With the business data text area, you can list other process or task data that is separated with a comma (,).

   For example,

   requisition.requester,currentPosition.jobTitle,currentPosition.replacement.lastName
   

   The field name of each of these variables in the Process Mining data source starts with 'tsk'.

   For example,

   tsk.requisition.requester, tsk.currentPosition.jobTitle, tsk.currentPosition.replacement.lastName
   

   After you choose what to include, click Next to upload your .csv raw data that is extracted by using the BAW connector.

   The following image is from a BAW application for reference on where the task data and exposed variables are sourced from.

   

Date type values from BAW are in the ISO8601 format, for example, 2011-08-04T03:12:53Z.

To work with IBM Process Mining NextGen, the format is changed to “%Y-%m-%d %H:%M:%S” through a function within BAW_process_app.py, which is the logic file within your BAW Process App. The format_dates function finds the columns and updates their values by filtering the names containing date or Time. The search is case sensitive.

If your results include any Date type columns that do not match the date or Time names, you might need to edit the Process App logic file. The columns can come from tracked or untracked variables.

The Process App logic file has a list of substrings that the columns are checked for:

def format_dates(event_list: pl.DataFrame) -> pl.DataFrame:
    #case sensitive list of substrings to watch out for in column names
    columns_search_text = [ "date", "Time"]

You must add the columns_search_text substring to the file. Use the affected column name or a substring of that name if you need to apply the date formatting to multiple columns.

5. Loading your data

   Drag and drop your .gz file or click upload to see the Folder window. Go to the file path in which you generated the zip file that contains the extracted data from BAW connector/extractor and upload it here.

   

The Process App then validates the data to validate that the data can be uploaded. After Process App completes validating the data, click Next to go to the Generate your process step.

6. Generating your process.

   In Generate your process step, IBM Process Mining generates a process based on the data that you upload in the Load your data step. You can see the progress of the generated process in any of the following statuses: PENDING, IN PROGRESS, completed, and ERROR. Each of the statuses is applicable under the following stages in the Generate your process step:

   • Transforming your data

   • Generating event log

   • Applying default analysis

   • Finalizing the process

   

   Click Finish to go to the IBM Process Mining Processes dashboard. The process generation runs in the background. When the Process App completes the process generation, you can see the new process that is created in the process list on the Processes dashboard.

   If you stay on the same window to view the progress of the process generation till the end, you see the status of each of the stages that are marked as completed.

   Click Finish to view your process on the process model tab.

Viewing the event log configuration

You can view the event log configurations for the process-mining project that you created by using the BAW business module Process App. To view the user input or the event log configurations, complete the following steps:

1. Go to Home page > All projects.
2. Select the process mining project that you created by using the process app.
3. Go to Data & Settings > Process App to view the event log configurations in the User inputs (Event logs configurations) section.